🎓 Mathematical Engineer | 💻 Computational Scientist
I'm a Master's student in Mathematical Engineering at Politecnico di Milano, with a passion for blending numerical methods for PDEs and machine learning. My thesis focuses on accelerating physical simulations using neural networks. With experience in Python, C++, and frameworks like PyTorch and SOFA, I strive to apply cutting-edge computational techniques to real-world problems.
Solving PDEs with Graph Neural Networks 🧠📊
A Python-based library to solve Partial Differential Equations (PDEs) using Graph Neural Networks (GNNs). This project offers a comprehensive pipeline for mesh generation, GNN training, and evaluation, inspired by recent advancements in computational mathematics.
Tech Stack: Python, FEniCS, PyTorch, NumPy
Accelerating Numerical Simulations 🏎️✨
An experimental space for working with the SOFA Framework, this project focuses on accelerating numerical simulations using various ML techniques.
Tech Stack: SOFA Framework, Python, PyTorch
Course Homework ⚙️🧬
A collection of projects addressing computational challenges in scientific computing and numerical analysis. This repository showcases my solutions created with the help of advanced C++ techniques.
Other challenges can be found in the APC repository, also in C++.
Tech Stack: C++
- Conducted research on accelerating numerical simulations with neural networks.
- Developed novel methods combining PyTorch and SOFA Framework.
- Presented results in team meetings and collaborated with supervisors.
- Experimenting with AI for Scientific Computing
- Exploring advanced applications of Graph Neural Networks
- Simulating soft robotics using SOFA Framework
- 📧 Email: andrea (dot) bonifacio [dot] 000 {at} gmail (dot) com
⭐️ Feel free to explore my repositories and connect!